From Pixels to Words -- Towards Native Vision-Language Primitives at Scale
Haiwen Diao
Paranioar
AI & ML interests
Vision-and-Language, Parameter-efficient Transfer Learning, Multi-modal Large Language Model
Recent Activity
updated
a dataset
about 7 hours ago
Paranioar/datacomp
published
a dataset
about 12 hours ago
Paranioar/datacomp
authored
a paper
11 days ago
The Prism Hypothesis: Harmonizing Semantic and Pixel Representations via Unified Autoencoding
Organizations
models
6
Paranioar/NEO1_0-2B-PT
Image-Text-to-Text
•
3B
•
Updated
•
8
•
1
Paranioar/NEO1_0-2B-MT
Image-Text-to-Text
•
3B
•
Updated
•
9
Paranioar/NEO1_0-2B-SFT
Image-Text-to-Text
•
3B
•
Updated
•
135
•
8
Paranioar/NEO1_0-9B-PT
Image-Text-to-Text
•
10B
•
Updated
•
4
Paranioar/NEO1_0-9B-MT
Image-Text-to-Text
•
10B
•
Updated
•
13
Paranioar/NEO1_0-9B-SFT
Image-Text-to-Text
•
10B
•
Updated
•
63
•
5